Good The Future Of Autonomous Vehicles Research Paper Example

After the turn of the last century, a tremendous economic upturn occurred, resulting from massive developments of a wide range of different technologies. As time has gone on, those advancing technologies continually affect the way we view our environment, and the ways in which people and nations interact. One particular innovative invention saw the light of day in 2011 when the state of Nevada passed a law facilitating the testing of driverless or autonomous cars, on the public highways system. Many found the occasion to be a cause for celebration, while others expressed hostility based on concerns about the potentially adverse effects on the safety of the general public.
Automation is a concept that is promoted and supported by automakers like Tesla Motors, Volkswagen, BMW, and the software giant, Google. These companies have been able to develop technologies that allow passengers to have a minimum amount of control over their car. Unfortunately the safe implementation of these technologies is still a few years from fruition and requires many legal and technological advances before the vehicles can be brought to market.
The federal government has only just begun developing laws and regulations that will be used in the management and regulation of the driverless cars. There are many concerns about the way driverless cars will operate. It is vital that we understand the processes of such advanced technology, and the ways they will be able to enhance the many aspects of the automotive industry, and in turn have a positive effect on our daily lives.
Self-driving cars are a fairly recent advancement in technology that has the potential to improve an individual's way of life in many distinct ways. The most obvious of course, is the automation of a task that in 2012, wasted on average 38 hours of production per person due to traffic congestion. According to TTI's 2012 Urban Mobility Report, those 38 hours cost every commuter nearly $820 per year, which amounts to more than $120 billion dollars wasted in 2012 alone. $120 billion dollars is a lot of money, in fact, it is just enough to purchase over 900,000 Model S sedans produced by Tesla Motors. The Model S is currently the only car on the market that claims a near autonomous driving experience. This is important because it brings consumers one step closer to rescuing all those wasted hours. Despite the intuitive use of technologies in place to control modern vehicles, the outlook for the widespread and safe use of driverless cars is still extremely dependent upon the development of more advanced artificial intelligence (AI).
Asimov’s claims began to bear fruition in the 1980’s when Carnegie Mellon’s Robotics department transformed a Chevy van into the first Navlab robot car. These robots were only mildly successful at driving themselves and were a very long wayfrom consumer production. The problem that existed with the Navlab cars is much the same that has hindered further advancements in this particular field. The computing equipment needed to operate the vehicles was so large that full size vans were necessary to haul all of it, completely negating any commercial use. On top of the physical requirements needed to operate such vehicles, there is an overwhelming concern about the security involved with such an open form of technology.
Security is a must when it comes to this kind of technology, taking into account that it will be very difficult (and unwise) to depend entirely on the logic, and the lack of moral standards present in self-aware vehicles. The safety of a ride will sometimes depend on the logic and the reasoning of the driver (Francis 82). There is also the issue of communication between the passenger and the car. It will be difficult to communicate with a car that lacks physical controls because the “brains” of a car will take instructions only from the people who operate it. (Luettel, Himmelsbach, and Wuensche 882). It is difficult; therefore, to imagine how the car will be corrected if it makes a mistake. For example, if a person was driving at high speeds and a child suddenly walks into the street it would be complicated if not impossible for a person to take control of an Autonomous Vehicle (or AV for short), and successfully avoid a collision. Society regards AI as more of a safety concern than a helpful innovation, because no one is sure of the capabilities of technology which is so new. We are trained to trust our kin over alien intelligence in most situations. Having a full dependency on machines without understanding the logistics can be a terrifying proposition for most people.
Before driverless cars will be allowed to operate on public roads, society will need appropriate answers to many different questions. For example, will it be possible to indicate that a passenger wishes to change their destination, and how will that be communicated to the car? Will drastic changes in the environment affect the stability and security of the car? If there are changes on roads, like new traffic lights that were installed overnight, would an AV be able to recognize and understand this type of situation, and in the event that its systems detect anomalies that it does not know how to deal with, what will be an appropriate reaction that will ensure the continued safety of everyone in the environment? There are some assumptions that human drivers make, that are completely out of context with logical behavior. For example a jaywalker crossing the street before the light indicates that it is safe to do so, may signal to an approaching car that they are leaving the confinement of the sidewalk. In a normal situation this would usually result in the driver slowing down and waiting for the pedestrian to pass. A driverless car may not be able to respond to such complex human behaviors. Unless there is a way in which the car will sense the pedestrian’s intentions on a somewhat intelligent level, the future of AVs looks relatively bleak, and will continue to be regarded as particularly dangerous.
Another major safety concern with allowing self-driving cars on public roads is that it is feared that criminals could use autonomous cars as self-propelled bombs. Criminals would hypothetically be able to control a car from a remote location, possibly circumventing driving laws that are in place to prevent dangerous situations. Although the driverless cars will open ways for mobility, they will also expose the general public to new forms of cyber security threats and potential terrorism. While being able to cause bodily harm is a very serious subject, it is also important to note that not all forms of harm could be directly influenced by man.
Weather is something that all Washingtonians are familiar with. Some would even go as far as calling Washington the rainy state. One of the driest states in the country is Nevada (World Media Group). The environments in the United States are neither uniform nor constant and have a huge range of changes every year. In order for autonomous vehicles to be truly effective, copious amounts of research would be imperative in order to accommodate the different climates in the country, and eventually the entire world. Even after all of this data has been accumulated, changes will still occur in global climates. In order for driverless cars to maintain a high level of safety, they would need to be able to perform exhaustive operations faster than the blink of an eye, otherwise it would be nigh on impossible to predict the cumulative variations in different weather conditions. Fortunately, many strides have been made in the world of mobile computing, that will allow for more powerful and robust operations to take place inside of a moving car.
Supercomputing is one aspect of the technology world that has always belonged to the elite, usually in the form of a University-funded project, or government-backed data centers. Recently, a computer graphics company called NVIDIA announced that they were producing a new kind of mobile computing chip that draws just ten watts of power, and is capable of one trillion floating operations per second. For the autonomous car industry, this means that the computing equipment needed for operation has shrunk from filling an entire van to the size of the average thumbnail. This miniaturization makes it possible for seamless integration between a car's systems and its operator, which in this case is the Tegra X1 Mobile Superchip. Unfortunately, this kind of technology is still at the stage of being an emerging market, which means that practical, future applications are as yet still in development.
Although many automakers are currently pursuing autonomous driving capabilities, there are just a few states which permit the sale and use of said vehicles on public roads. Most of the complications in this regard arise from the complex task of applying current driving laws to cars that may be operated without a human occupant. It is probably legal to direct the controls of a motor vehicle by computer to steer it, as well as controlling acceleration and braking – without human input in real time (Smith, 2012). By implication, deciding whether or not driverless cars will be allowed into the mainstream transportation market may be a long and arduous ordeal.
In order for a person to operate a passenger car in Washington State, they are first required to obtain a valid driver’s license issued to state residents. Unfortunately, in the case of autonomous vehicles, operation falls solely in the “hands” of the computer chips residing in the car’s innards. If the actions of an autonomous vehicle are not controlled by human occupants, there may not be a need for driver’s licenses to be issued before using a driverless car. Currently there are no laws that inhibit computers from self-operation. However, vehicles may in the future be required to pass a standardized driving test. A practical alternative to this will most likely require automakers to include a form of control that may be employed by occupants. Another legal conundrum is that even though state laws are in place to make sure that automakers do not make oddly-shaped steering wheels, there appear to be no laws that require vehicles to actually have a steering wheel. For example, new testing regulations introduced in California for Google's driverless cars, require that in an emergency situation, drivers must take “immediate physical control” of a vehicle using the public roads system. That seems to require all driverless cars to have a steering wheel, and brake and accelerator pedals. Using the steering wheel as an example, it becomes very clear that laws keeping the wheel in the hands of a human driver will be difficult to enforce.
The decision of whether to allow driverless cars on the road is not in the hands of the federal government. States have the power to set driving laws, like the age requirement for a driver's license and the range of penalties for driving while talking on a cellphone. Taking advantage of this state power, California, Michigan, Nevada, and Florida have all legalized the use of fully autonomous cars in the last four years. In Nevada, driverless vehicles are specifically licensed one-by-one by the state with a special license plate and a requirement that at least two people, one in the driver's seat, be in the car at all times. For every other state in the U.S., driverless cars are not technically illegal because no law says cars must have drivers. This means that not only states but also the federal government must draft appropriate regulations which will permit the operation of a self-driving vehicle by any car owner when such vehicles become generally available.
Notwithstanding these legal and other issues surrounding this new technology, driverless vehicles and prototypes are being built and tested by various automakers, eager to gain a foothold into what could be a simply enormous market for the successful manufacturer(s). And this is the situation not just in the United States, but in other countries, too. For example, the Dutch government has just authorized a local service using driverless shuttles in the city of Wageningen, commencing in December this year. These electric vehicles will transport up to 8 passengers from the railway station to the university over circa 6 kilometers of public roads at a maximum design speed of 50 kilometers per hour. Even though the service will be in the nature of tests, the vehicles will make the journeys autonomously (i.e. without safety drivers aboard), monitoring the operations remotely. Before the service can begin, amendments to Dutch traffic regulations will need to be approved by both chambers of the Dutch parliament. Provided there are no unforeseen obstacles, the first fully autonomous shuttle service (without a human backup driver) in the world could be operating in the Netherlands by the end of 2015 (“Netherlands first to operate a self-driving shuttle in public traffic?” 2015).
Another and perhaps even more exciting and ambitious example of projects involving driverless vehicles is the next generation Audi A8 model. Audi has announced that it will have the capability to “drive with full autonomy.” The company has the ambition to be the first major car manufacturer to offer a driverless car to the mass market. The car will have cameras and use LIDAR (light detection and ranging) technology, and according to Audi will drive more safely than if driven by humans. The systems will use a computer architecture in which all processing will be performed by a minimum of two independent computers. Although the aim is to have the car available by 2017, the company cautioned that delays to this projected schedule could arise due to current legal obstacles associated with autonomous driving (Hars, Nov. 2014).
A further cautionary note is appropriate here. The lidar system intended as the “eyes” for the proposed new Audi and other car makers is currently very costly. According to Shchetko (2014), the cost per vehicle is for an advanced system is currently between $30,000 and $85,000, although some prototype vehicles are utilizing as many as four smaller and cheaper units. The article reports that the system used on the Google driverless prototypes costs over $75,000. Although manufacturers predict those prices can be brought down, research has not yet identified just when the price will be low enough to allow the system to be affordable for a mass market vehicle. A target price of $500 is thought to be the point at which the system will be viable for mass production. The lidar systems offer high resolution “images” of objects detected, and in three dimensions, too. According to the Ford Motor Company, it can tell the difference between (say) a cat and a paper bag at a range of circa 100 yards. Most prototype vehicles have the lidar units mounted on top of the roof, which provides good, all-round vision, but for the mass market it is likely that units would be mounted more discreetly, such as behind the grille. However, that would mean additional lidar sensors would be needed in other locations on the vehicle. Furthermore, because the lidar system is color-blind, the vehicles need cameras in addition, to be able to “read” traffic lights, etc (Shchetko, 2014).
A possible alternative solution to the high cost of the currently available lidar systems may have been found by Sony. From this year (2015), they will offer a new automotive sensor that they claim performs well in low light and can even provide color images in moonlight. The article does suggest though that it is unlikely that fully autonomous driverless cars used in urban environments will be able to dispense with lidar sensors for at least the coming few years (Hars, Sept. 2014).
Returning to the subject of amending rules and regulations to allow the use on public roads of driverless vehicles, a major problem for the vehicle manufacturers who wish to export their products is that the applicable laws vary from country to country. In an attempt to overcome this obvious difficulty, the United Nations Forum WP29 is attempting to harmonize those national vehicle laws and regulations using a series of informal working groups. In order to keep pace with the rapid evolution of the technology for autonomous vehicles, the group dealing with Intelligent Transport Systems (ITS) has been redefined as the ITS / Automated Driving Group and its focus has been shifted accordingly. Members of that group are now establishing the basis for future regulatory legislation (“Global technical regulations for autonomous vehicles: Informal working group established” 2014).
As regards how policy makers should approach the subject of driverless cars and the associated and rapidly-evolving technology, Hars (Oct. 2014) provides his own interpretative expansion of the five guiding principles for policy makers proposed by Marc Scribner, a research fellow and specialist in telecommunications and transportation policies at the Competitive Enterprise Institute. The principles (as expanded and interpreted by Hars) can be summarized as follows:

Recognize, support and encourage the enormous potential of driverless vehicles:

Policy makers should become familiar with the possible benefits of autonomous vehicles. Firstly, the concepts must be clearly defined, distinguishing driverless cars (which can drive on their own, even with no human on board) from associated technologies including systems providing driver assistance. As well as offering the probability of drastically reduced numbers of accidents, driverless cars also provide new opportunities of mobility for individuals not holding a driver’s license – including the disabled and elderly. In addition they use less energy, including using alternative fuels, and reduce road congestion. Policy makers must also recognize that driverless cars can mitigate or obviate various longstanding issues. Politicians must promote and support the technology right now. Waiting to see what happens is not a viable approach, although the associated technological risks must nonetheless be given due consideration (Hars, Oct. 2014).

Turn away from the principle of precaution:

For driverless cars, safety is both a major concern and an important benefit. Driverless cars will drive far safer than human drivers. Their safety equipment includes all-round “vision” sensors which are constantly alert, are not subject to tiredness or fatigue or impairment through drugs or alcohol, and always adopt a driving strategy that is defensive in nature and always minimizes risks. However, making that decision to allow the first driverless cars to drive unsupervised on public roads is difficult: what if something fails? The precautionary principle overcomes this quandary by requiring designers to prove that cannot happen, which is very difficult to do. Hence, applying that principle rigorously could introduce serious delays into a driverless car program. Unfortunately, delay can be costly in human lives given that the annual death toll from traffic accidents in the US alone is 33,000 and over one million globally. For that reason alone, a way forward should be found - not an easy task for our policy makers but one which can save lives (Hars, Oct. 2014).

Don’t second-guess the evolution of the law and the technology:

Because there are so many ways this technology can develop, coupled with so many possible areas of evolution in business and in society, it is difficult if not impossible to predict how the technology and consequently the law will evolve. It is foolish and even dangerous to assume any one future trend, then legislate for that scenario only to find subsequently that the reality is quite different. Possible issues of uncertainty include whether the driverless car technology will evolve progressively from existing driver assistance systems. Also, will the US be the first nation to legalize the use of autonomous vehicles? And will a number of European nations be inhibited from adopting driverless cars by the road traffic requirements of the Vienna Convention? (Hars, Oct. 2014).
4. Allow the innovators to be innovative:
In this area Hars admits to differing with Scribner’s views which proposed to “minimize legislative and regulatory intervention.” Hars believes that because transportation legislation is predominantly based on the human driver concept, there are many existing laws that will require amendment to allow progress in this new and safer area of transportation technology. Otherwise, innovators will find progress difficult if not impossible. And action in that area should begin immediately – before this new generation of road transport is available for public use. It must be emphasized that driverless cars dramatically change the fundamental concept of a car (Hars, Oct. 2014).
5. Preserve the neutrality of the technology:
As far as possible any amended or new legislation and regulation should be framed so as to be technologically neutral. That means avoiding the tendency to specify and therefore favor any one particular technical approach (Hars, Oct. 2014).
In conclusion, it is evident that driverless cars have a bright future and once in widespread use can save lives, due the increased safety inherent in their design and their built-in measures to prevent collisions, etc. However, there are many hurdles to overcome before they can realistically become affordable and available. There are technological issues and cost reduction issues yet to be resolved, as well as legal and regulatory changes that will need to implemented – possibly including a global rationalization of road traffic regulations to facilitate opportunities for international sales of these autonomous vehicles.

Works Cited:

Asimov, Isaac. “Visit to the World's Fair of 2014.” Editorial. New York Times 16 Aug. 1964: (n. pag.). Web. Accessed 8 Feb. 2015.
Bogost, Ian. “The Secret History of the Robot Car.” Atlantic 314.4 (2014): 93. ATLA Religion Database. Web. Accessed 6 Feb. 2015.
Francis, Esther. “Driverless Cars in 2022 Making the Vision a Reality.” Automotive Industries 191.1 (2012): 32-33. Web. Accessed 6 Feb. 2015.
“Global technical regulations for autonomous vehicles: Informal working group established.” Driverless car market watch. 30 Dec. 2014. Web. Accessed 3 March 2015.
Hars, Alexander. “First fully autonomous Audi expected by 2017.” Driverless car market watch. 22 Nov. 2014. Web. Accessed 4 March 2015.
Hars, Alexander. “Five guiding principles for autonomous vehicle policy.” Driverless car market watch. 20 Oct. 2014. Web. Accessed 4 March 2015.
Hars, Alexander. “Sony enters the market for automotive imaging sensors.” Driverless car market watch. 1 Sept. 2014. Web. Accessed 4 March 2015.
Luettel, Thorsten, Michael Himmelsbach, and H-J. Wuensche. “Autonomous ground vehicles—concepts and a path to the future.” Proceedings of the IEEE100.Special Centennial Issue (2012): 1831-1839. Web. Accessed 11 Feb. 2015.
“Netherlands first to operate a self-driving shuttle in public traffic?” Driverless car market watch. 9 Feb.2015. Web. Accessed 3 March 2015.
“NVIDIA Launches Tegra X1 Mobile Super Chip.” NVIDIA Newsroom, 4 Jan. 2015. Web. Accessed 11 Feb. 2015.
“Passenger Car.” RCW 46.04.382. Washington State Legislature, 1963. Web. Accessed 20 Feb. 2015.
Porikli, Fatih and Van Gool, Luc. “Special issue on car navigation and vehicle systems.” Machine Vision and Applications 25.3 (2014): 545-546. Web. Accessed 11 Feb. 2015.
Riener, Andreas. “Who cares about trust, grade of traveling & quality of user experience in a world of autonomous cars?” Proceedings of the 6th International Conference on Automotive User Interfaces and Interactive Vehicular Applications. ACM, 2014. Web. Accessed 11 Feb. 2015.
Ros, German, et al. “Visual slam for driverless cars: A brief survey.” Intelligent Vehicles Symposium (IV) Workshops. 2012. Web. Accessed 10 Feb. 2015
Shchetko, Nick. “Laser Eyes Pose Price Hurdle for Driverless Cars.” The Wall Street Journal. 21 July 2014. Web. Accessed 4 March 2015.
Seshan, Jayaraman, & Sandhya Maitra. “Efficient Route Finding and Sensors for Collision Detection in Google’s Driverless Car.” (2014). International Journal of Computer Science and Mobile Computing Vol.3 Issue.12, December- 2014, pg. 70-78. Web. Accessed 12 Feb. 2015.
Schrank, David, Eisele, Bill, & Lomax, Tim. “TTI's 2012 Urban Mobility Report.” Urban Mobility Information. Texas A&M Transportation Institute The Texas A&M University System, (n.d.). Web. Accessed 07 Feb. 2015.
Smith, Bryant W. “Automated Vehicles Are Probably Legal in the United States.” Stanford Law School, 1 Nov. 2012. Web. Accessed 20 Feb. 2015.
“Tesla Model S Pricing.” Model S Design Studio. Tesla Motors, 2015. Web. Accessed 6 Feb. 2015.
“U.S. Average Precipitation State Rank.” World Media Group, 2015. Web. Accessed 11 Feb. 2015.